Rudy Lai

AI @ CSX

Freight rail, eastern U.S.
Industry
Last updated
July 3, 2025 at 10:44 AM

Summary

  • CSX has progressively adopted AI and machine learning technologies since 2016, starting with IoT-enabled machine learning for train delay indexing and operational insights, advancing to AI-powered safety tools like trespassing detection developed by Rutgers researchers by 2022.
  • From 2024 onward, CSX significantly scaled AI integration with initiatives such as deploying the AI assistant 'Chessie' through Microsoft Copilot Studio and Azure AI Foundry, engaging over 1,000 customers in 4,000+ interactions within 45 days, highlighting rapid commercialization and enhanced customer experience in rail logistics management.
  • Recent efforts up to Q3 2025 include cloud-native AI solutions for real-time analytics reducing derailments, advanced sensor and camera AI systems for safety improvements, and plans for expanding AI-driven multi-agent orchestration, reflecting a maturing AI strategy aimed at operational excellence, risk reduction, and superior customer engagement.

VIBE METER

More AI announcements = more VIBE
Q1 23Q2 23Q3 23Q4 23Q1 24Q2 24Q3 24Q4 24Q1 25Q2 25Q3 25🔥🔥🔥🔥

5 AI Use Cases at CSX

Safety Monitoring
2025
Traditional
Generative
Agentic
Outcome
Risk
Integration of AI-powered camera technology and edge computing enables real-time hazard detection on railroads, enhancing situational awareness and preventing accidents. [1][2]
Customer Assistance
2025
Customer Facing
Traditional
Generative
Agentic
Outcome
CSX deployed a generative AI chatbot named 'Chessie' integrated within its ShipCSX portal using Microsoft Copilot Studio to automate freight tracking, shipment management, and customer inquiries, significantly improving customer experience and operational efficiency. [1][2]
Operational Analytics
2025
Traditional
Generative
Agentic
Outcome
Risk
By leveraging cloud-based AI solutions on Microsoft Azure, CSX achieves real-time data streaming and predictive analytics that reduce derailments and optimize rail operations. [1]
Trespassing Detection
2022
Traditional
Generative
Agentic
Outcome
Risk
Rutgers and CSX utilize AI-aided video analytics and surveillance to detect railroad trespassing events, preventing accidents and fatalities at crossings through early warnings and improved monitoring. [1][2][3]
Delay Prediction
2016
Traditional
Generative
Agentic
Outcome
Costs
CSX uses IoT-enabled machine learning models to create train delay indexes that quantify trip failures and their associated costs, enhancing scheduling and operational planning. [1]

Timeline

2025 Q3

3 updates

Ongoing research and applications include advanced AI for railroad trespassing detection analyzing thousands of hours of footage, edge AI for bridge impact detection, and safety improvements integrating AI with sensors and camera technologies.

2025 Q2

2 updates

CSX deployed Microsoft Azure cloud and AI solutions to transform rail operations, launching 'Chessie,' a generative AI assistant integrated into the ShipCSX portal, achieving over 1,000 customer engagements in 45 days and significantly enhancing freight tracking and shipment management.

2025 Q1

4 updates

Multiple initiatives highlighted: UNM's research on neuromorphic sensors for rail maintenance, BNSF's AI efforts for maintenance and yard checks, and growing recognition of generative AI's value in freight railroads.

2024 Q4

1 updates

Railroads ramped up AI use in transportation planning and operational adjustments to meet dynamic demands, demonstrating increased AI integration into logistics.

2024 Q3

2 updates

Wi-Tronix deployed AI-powered camera tech to detect hazards and improve railroad safety, alongside government research on AI intruder detection systems.

2024 Q2

1 updates

Federal Railroad Administration published research on building a railroad trespassing database using AI from a Rutgers-led project, reinforcing safety initiatives.

2024 Q1

2 updates

CSX introduced an AI-powered chatbot to streamline real estate inquiries, enhancing customer engagement and self-service capabilities, alongside growing industry reflections on AI for railway operations efficiency.

2023 Q4: no updates

2023 Q3

1 updates

Industry-wide discussions on AI's concept and subset machine learning highlighted its growing adoption in transportation, underscoring technological awareness in railroads.

2023 Q2: no updates

2023 Q1: no updates

2022 Q4: no updates

2022 Q3: no updates

2022 Q2

1 updates

Rutgers researchers developed an AI-aided railroad trespassing detection tool to enhance safety and reduce fatalities at crossings.

2022 Q1: no updates

2021 Q4: no updates

2021 Q3: no updates

2021 Q2: no updates

2021 Q1: no updates

2020 Q4: no updates

2020 Q3: no updates

2020 Q2: no updates

2020 Q1

1 updates

The rail industry, including CSX, began experiencing AI-driven transformation impacting workforce roles through automation, AI, and robotics.

2019 Q4: no updates

2019 Q3: no updates

2019 Q2: no updates

2019 Q1: no updates

2018 Q4: no updates

2018 Q3: no updates

2018 Q2: no updates

2018 Q1: no updates

2017 Q4: no updates

2017 Q3: no updates

2017 Q2: no updates

2017 Q1: no updates

2016 Q4: no updates

2016 Q3: no updates

2016 Q2

1 updates

CSX initiated AI adoption with IoT-enabled machine learning to analyze train delays, creating a train delay index for better operational cost insights.